Visual impairment and blindness (VI/B) are well recognised as significant contributors to poor quality of life and reduced productivity.1 The World Health Organization (WHO) estimated that in 2010, there were 39 million people blind and an additional 246 million people visually impaired.2 In 80% of these cases the cause of blindness (B) and visual impairment (VI) is believed to be either treatable or preventable.3 Consistent with evidence of other sensory disability research, VI/B disproportionately affect people living in low and middle-income countries (LMIC).2,4 The WHO estimates that the prevalence of blindness is approximately 0.3% in high-income countries and likely to be greater than 1% in LMIC.5 Furthermore, VI/B affects a greater proportion of women than men;6 a 2001 meta-analysis estimated that women accounted for 64.5% of all blind adults worldwide.7There is limited research and routine surveillance on eye health in Pacific Islands countries (PICs), resulting in fragmented understanding of ophthalmic epidemiology throughout the Pacific region.8 Where population surveys have been undertaken, they frequently only include adult populations, providing little or no information on childhood VI/B.9,10 PICs (Figure 1) have diverse cultures, religious affiliations, political histories, geography11,12 and resources.13 Despite their diversity, these countries share barriers to prosperity and economic success, including remoteness, political instability, lack of human resources and vulnerability to climate change and other natural disasters.11,14-16 Figure 1: Map of the South West Pacific Region Eye care is delivered within the PICs by national health systems and private organisations, such as Fred Hollows Foundation New Zealand (FHFNZ).17 Other providers have included Brien Holden Vision Institute,18 Marine Reach Ministries,19 the New Zealand Medical Treatment Scheme20 and the Royal Australasian College of Surgeons (RACS) Pacific Islands Project (PIP).21 Early population surveys indicate that cataract is the leading cause of blindness and one of the most common causes of VI in PICs.9,10,22 Surgery for this condition provides a fast, relatively inexpensive and low-risk method of restoring sight.23-26Opthalmology outreach services are necessary in parts of the Pacific region where geographic isolation, small populations and the lack of human and other resources mean that many people cannot access specialist health services. The standard model of outreach health service delivery in developing countries entails small teams of clinicians, both local and from overseas, working alongside eye-care nurses.Anecdotal reports from clinicians involved with FHFNZ outreach clinics suggested that there may be a decreasing number of patients presenting for assessment, and in particular a reduction in the number of patients requiring cataract surgery.Data collected during the outreach visits provide insight into the profile of presenting eye health problems in the region from the perspective of one provider and offer an opportunity to test whether there have been changes over time, and if so, what modifications may be required in the ophthalmic services in the region.Materials and methodsRoutine existing data (secondary data) collated at seven Pacific Island outreach clinics between 2009 and 2013 were analysed using Excel. Outreach clinics are defined as any clinic held outside the premises of designated program centers. Countries included in this analysis were Fiji, Kiribati, Papua New Guinea (PNG), Samoa, the Solomon Islands, Tonga and Vanuatu. Data were accessed with permission from a regional eye health provider database as Excel spreadsheets and written reports. Raw data were de-identified to preserve the anonymity of the patients. This project was exempt from requiring ethical approval by The University of Auckland Human Participants Ethics Committee (UAHPEC), as this study utilised secondary data collected by the FHFNZ and did not involve contact with human participants. This project was also released from requiring formal ethics approval by the Health and Disability Ethics Committee. The study adheres to the tenets of the Declaration of Helsinki in that the rights of all human participants are protected through the process of de-identification of data and ethical consideration in the analysis and reporting of data.26Diagnosis was coded as per the WHO guidelines for primary diagnosis,27 as the current outreach data do not allow for clear identification of the presenting issue or diagnosis, as opposed to concurrent ophthalmic comorbidities. Additionally, all cases of trachoma and trichiasis were recoded, even in cases where this was not the primary diagnosis. Where Excel spreadsheets were not available, statistics were sought from written reports. Data were aggregated by country and year. Descriptive statistical analysis was then applied.ResultsIn total, 67 outreach clinics were conducted in seven countries between 2009 and 2013.Outreach data indicate that the mean number of patients seen at outreach clinics appear to be decreasing in Fiji, Samoa and the Solomon Islands (Figure 2). Data were not available for outreach clinics held in Kiribati in 2012 or the Solomon Islands in 2011. Data from PNG stand out for the high mean number of patients per clinic even after accounting for the greater length of PNG outreach clinics.Figure 2: Mean number of patients assessed at outreach clinics by location and year *No data available for clinics held in Kiribati 2012 and the Solomon Islands 2011.Additionally, the number of patients reviewed at PNG outreach clinics appears to be increasing, in contrast to other outreach locations. Notably, outreach clinics in PNG are run differently to those in other locations. PNG does not have locally-based Pacific Eye Institute trained ophthalmic nurses to undertake pre-outreach screening of patients. Therefore, ophthalmic nurses conduct screening simultaneously with medically run outreach clinics and additionally, clinics last for two weeks in PNG, as opposed to one week in all other locations.The percentage of patients whose primary diagnosis is cataract is shown in Table 1. According to these figures, cataract typically contributes to between 40% and 70% of all primary diagnoses at outreach clinics. However, data from PNG in 2013 indicate that the proportion of patients with cataract as a primary diagnosis was considerably lower only 14.1% of all patients reviewed. In contrast, clinics held in Fiji in 2013 had an unusually high proportion of cataract diagnoses. In this year, 94.2% of all patients seen at outreach clinics in Fiji had cataract disease as their primary diagnosis.Table 1: Percentage of outreach patients with cataract as primary diagnosis Clinic locations: 2009 2010 2011 2012 2013 Fiji 65.1 45.4 94.2 Kiribati 59.4 Papua New Guinea 14.1 Samoa 69.2 62.3 60.0 Solomon Islands 64.5 65.1 Tonga 54.9 55.1 Vanuatu 41.1 53.4 48.8 Mean 58.5 56.1 56.7 Note: No data available for any outreach sites from 2009 and 2010.Table 2 illustrates the number of surgeries and the proportion of cataract surgeries performed in each outreach country between 2009 and 2013, according to all available outreach data. This information was not available for all of the outreach clinics during this period. Consequently, this table likely under-represents the true number of surgeries performed at FHFNZ outreach clinics. The proportion of cataract surgeries ranged from as low as 59.0% (Vanuatu, 2009) to 97.1% of all surgeries (Fiji, 2013). It should be noted that the changes observed for Fiji may be related to changes in service provider. In addition, there were no data for PNG and Kiribati.Table 2: Cataract surgery: Annual totals and percentage of all surgeries (in brackets) by location and year Clinic locations: 2009 2010 2011 2012 2013 Total Fiji 89 % unknown 337 (89.6) 387 (92.3) 332 (95.4) 339 (97.1) 1,484 % unknown Kiribati 89 (92.7) 106 (85.5) 195 (88.6) Papua New Guinea 121 (83.4) 183 (85.9) 255 (90.0) 217 (71.0) 776 (82.1) Samoa 105 (82.0) 127 (81.4) 322 (78.5) 167 (85.1) 721 (81.1) Solomon Islands 127 (86.0) 156 (72.0) 126 (60.3) 283 (80.0) 692 (74.6) Tonga 74 (74.7) 207 (77.8) 281 (77.0) Vanuatu 120 (59.0) 188 (62.0) 93 (73.8) 50 (65.8) 116 (66.3) 567 (64.2) Total 209 % unknown 978 (88.8) 946 (83.6) 1,248 (82.1) 1,435 (81.2) 4,816 % unknown Table 3 presents the proportion of all cataract surgeries that were performed on female patients. Data indicate that at most outreach locations, less than half of all cataract surgeries were performed on females. Notably, in 2013 in PNG only 30.4% of all cataract surgeries were undertaken on females.Table 3: Percentage of patients who recieved cataract surgeries who are female Clinic locations: 2009 2010 2011 2012 2013 Fiji 53.9 58.4 63.3 Kiribati 59.8 35.7 Papua New Guinea 30.4 Samoa 45.3 49.0 49.7 Solomon Islands 46.8 45.1 Tonga 45.9 59.6 Vanuatu 48.8 35.7 42.5 Note: No data available for any outreach sites from 2009 and 2010. Table 4 presents the percentage of patients whose primary cause of VI or B is an uncorrected refractive error (URE). Available data indicate that refractive errors contribute a much greater proportion of cases of VI/B at outreach clinics in PNG compared to other locations. In 2013, 60.6% of all patients presenting to outreach clinics in PNG had a primary diagnosis of URE.Table 4: Percentage of outreach clinic patients with uncorrected refractive error as the primary diagnosis Clinic locations: 2009 2010 2011 2012 2013 Fiji 6.8 33.7 0 Kiribati 1.1 Papua New Guinea 60.6 Samoa 2.8 7.3 9.3 Solomon Islands 2.6 2.4 Tonga 0.4 1.7 Vanuatu 21.5 10.3 3.9 Note: No data available for any outreach sites from 2009 and 2010. The proportion of patients at outreach clinics whose primary diagnosis is diabetic retinopathy (DR) is displayed in Table 5. Data are only available from 2011 onwards. DR accounted for less than 4.0% of primary diagnoses at most locations between 2009 and 2011. However, data from Tongan outreach clinics showed a much greater proportion of DR cases: 26.2% in 2012 and 14.2% in 2013.Table 5: Percentage of outreach clinic patients with diabetic retinopathy as the primary diagnosis Clinic location: 2009 2010 2011 2012 2013 Fiji 1.4 0.2 0.3 Kiribati 0.6 Papua New Guinea 0.2 Samoa 3.2 3.7 0.7 Solomon Islands 1.1 0.6 Tonga 26.2 14.3 Vanuatu 0.8 1.7 1.2 Note: No data available for any outreach sites from 2009 and 2010. All patients who were identified as having trachoma or trichiasis in outreach data were recorded (Table 6). This included cases where trachoma or trichiasis did not qualify as the primary diagnosis. Information on the stage of trachoma infection was not available in outreach data. No cases of trachoma or trichiasis were identified in data collected at outreach clinics in Kiribati, Samoa or Tonga. A total of 51 patients with trachoma or trichiasis were identified between 2009 and 2013. The majority of these cases (35 people) were identified at a single outreach clinic in PNG. Notably, 90% of all cases in the Solomon Islands underwent surgical correction for trichiasis.Table 6: Trachoma cases by location: 2011 to 2013 Clinic locations: Number of patients Proportion of cases seen % Female Mean age % with BTR # Fiji 2 0.1 100.0 38.7 0 Kiribati 0 Papua New Guinea 35 51.4 0 Samoa 0 Solomon Islands 10 1.3 70.0 65.3 90.0 Tonga 0 Vanuatu 4 0.6 50.0 69.3 75.0 * Fraction of clinics where trachoma was specifically recorded in outreach reports or where raw data was available for analysis.# Bilamellar tarsal rotation surgical treatment of trichiasis. DiscussionPreventable vision loss remains a persistent public health concern in the Pacific Islands region. Contextual factors, including the vast geographical isolation, chronic shortage of eye nurses and ophthalmologists and, to some extent, a weak primary health sector, contribute to the burden faced by the region.Detailed studies of the epidemiology of eye conditions have been carried out in a small number of PICs, but these are costly and alternative approaches need to be considered. Analysis of routinely collected clinical data cannot replicate the full understanding gained from dedicated population-based eye health surveys. However, as this paper illustrates, using the activities of a single eye health provider can be an efficient means to capture the profile of presenting cases, to understand in broad terms patient demography and to identify hot spots of infectious diseases, for example, trachoma. This method is particularly promising when there is an agreement between local health systems and NGO eye health providers to share data.At present, it is difficult to confidently establish how many organisations are providing eye health services in PICs, the quality of these services, and what impact services are having on the burden of eye disease in the Pacific.This situation is exacerbated by the paucity of detailed ophthalmic epidemiology data from Pacific Island countries. Moreover, maintaining responsive databases is prohibitively expensive and must compete with other health care demands.Data generated at outreach clinics may provide an on-going flow of information to support surveillance of eye health in the region. These records can be used for clinical surveillance to monitor the changes seen in disease presentations over time and across locations, and in this way assist eye health teams to evaluate their work and assist with planning and the prioritisation of resources for outreach services. Outreach eye health services tend to be provided by regional NGOs or by sporadic visiting international medical teams, possibly leading to minimal follow-up and no clear accountability to national or regional agencies. Therefore, linking the various individual and NGO eye health service data to national (and regional) health information systems may greatly enhance data quality.A commitment to the collection and open reporting of outreach data by eye health service providers in the Pacific will allow more accurate prediction of emerging trends in the ophthalmic health of PICs. This is of great importance for monitoring the changes in disease presentations, particularly as PICs transition toward a greater burden of non-communicable diseases.27-30 It is likely in the future there will be greater numbers of patients presenting with cataract, DR31,32 and hypertensive retinopathy.33,39 Eye health teams can also contribute key data to determine access to eye health services for vulnerable groups, such as women, children, elderly and ethnic minority groups. This information provides in-country government agencies and visiting ophthalmology teams a clearer profile of groups with poorer access to eye health services and supports the development of strategies to address disparities.Data collected from Pacific outreach eye health clinics between 2009 and 2013 indicate presentations at outreach clinics decreased in three out of the seven counties surveyed (Fiji, Samoa and the Solomon Islands). Several explanations are viable here: a decline in the number of patients due to a diminishing backlog of cataract-induced VI/B; changes to the number and/or type of clinicians at outreach clinics; or cataract prevalence may remain high, but patients are not presenting to outreach clinics for treatment. Alternatively, the changing disease profile in the region may mean patients are not presenting as their conditions are largely asymptomatic, undermining the reason and certainly the urgency of presenting to an outreach clinic.Of note, our data also showed high numbers of URE cases were reported in PNG a condition associated with poor access to primary eye health services. Tongan clinics are notable for the high number of DR cases reported, consistent with the high prevalence of diabetes in that country.28 Also, our findings indicate that between 2009 and 2013, in most locations, more men than women underwent cataract surgery, although most studies find a higher prevalence of cataract amongst women in developing countries.28-32Likewise, clinical surveillance is critical for providing an estimate of the burden of trachoma in PICs in the absence of up-to-date survey data on the prevalence and effects of trachoma in PICs. Collaborative use of clinical data from outreach clinics could provide an insight into where trachoma infections are prevalent or where a greater proportion of infections are progressing to VI/B. Eye health teams should also consider using negative reporting to ensure that all cases of trachoma are identified, given the drive to see blinding trachoma eliminated globally34-36 and the current lack of information on its prevalence and effects in the Pacific.22,35Our analysis of outreach data provides evidence of the importance of investing in robust and reliable data collection systems that prioritise accountability and transparency. We argue that bold innovation and collaboration in data collection, reporting and sharing mechanisms are essential in resource scarce settings. In the absence of this, we will continue to be prone to speculation about the shifting burden of eye health and disease in the region.ConclusionAnalysis of one eye health service providers data from outreach clinics across seven PICs from 2009 to 2013 indicates a reduction in the number of patients presenting for treatment. This trend, and others reported here, should be interpreted cautiously due to concern about data completeness and quality. Also, these data relate to those who present after (in most countries) screening by the eye care nurses. We have no information on those who may suffer from eye health problems, but do not present to clinics.A commitment to quality data management and surveillance strategies for all eye health outreach programmes is vital to ensure resources are reaching those who need it, that health equity lies at the core of the efforts to deliver eye health services and finally, that innovation and collaboration are essential components of Pacific regional eye health initiatives.

Visual impairment and blindness (VI/B) are well recognised as significant contributors to poor quality of life and reduced productivity.1 The World Health Organization (WHO) estimated that in 2010, there were 39 million people blind and an additional 246 million people visually impaired.2 In 80% of these cases the cause of blindness (B) and visual impairment (VI) is believed to be either treatable or preventable.3 Consistent with evidence of other sensory disability research, VI/B disproportionately affect people living in low and middle-income countries (LMIC).2,4 The WHO estimates that the prevalence of blindness is approximately 0.3% in high-income countries and likely to be greater than 1% in LMIC.5 Furthermore, VI/B affects a greater proportion of women than men;6 a 2001 meta-analysis estimated that women accounted for 64.5% of all blind adults worldwide.7There is limited research and routine surveillance on eye health in Pacific Islands countries (PICs), resulting in fragmented understanding of ophthalmic epidemiology throughout the Pacific region.8 Where population surveys have been undertaken, they frequently only include adult populations, providing little or no information on childhood VI/B.9,10 PICs (Figure 1) have diverse cultures, religious affiliations, political histories, geography11,12 and resources.13 Despite their diversity, these countries share barriers to prosperity and economic success, including remoteness, political instability, lack of human resources and vulnerability to climate change and other natural disasters.11,14-16 Figure 1: Map of the South West Pacific Region Eye care is delivered within the PICs by national health systems and private organisations, such as Fred Hollows Foundation New Zealand (FHFNZ).17 Other providers have included Brien Holden Vision Institute,18 Marine Reach Ministries,19 the New Zealand Medical Treatment Scheme20 and the Royal Australasian College of Surgeons (RACS) Pacific Islands Project (PIP).21 Early population surveys indicate that cataract is the leading cause of blindness and one of the most common causes of VI in PICs.9,10,22 Surgery for this condition provides a fast, relatively inexpensive and low-risk method of restoring sight.23-26Opthalmology outreach services are necessary in parts of the Pacific region where geographic isolation, small populations and the lack of human and other resources mean that many people cannot access specialist health services. The standard model of outreach health service delivery in developing countries entails small teams of clinicians, both local and from overseas, working alongside eye-care nurses.Anecdotal reports from clinicians involved with FHFNZ outreach clinics suggested that there may be a decreasing number of patients presenting for assessment, and in particular a reduction in the number of patients requiring cataract surgery.Data collected during the outreach visits provide insight into the profile of presenting eye health problems in the region from the perspective of one provider and offer an opportunity to test whether there have been changes over time, and if so, what modifications may be required in the ophthalmic services in the region.Materials and methodsRoutine existing data (secondary data) collated at seven Pacific Island outreach clinics between 2009 and 2013 were analysed using Excel. Outreach clinics are defined as any clinic held outside the premises of designated program centers. Countries included in this analysis were Fiji, Kiribati, Papua New Guinea (PNG), Samoa, the Solomon Islands, Tonga and Vanuatu. Data were accessed with permission from a regional eye health provider database as Excel spreadsheets and written reports. Raw data were de-identified to preserve the anonymity of the patients. This project was exempt from requiring ethical approval by The University of Auckland Human Participants Ethics Committee (UAHPEC), as this study utilised secondary data collected by the FHFNZ and did not involve contact with human participants. This project was also released from requiring formal ethics approval by the Health and Disability Ethics Committee. The study adheres to the tenets of the Declaration of Helsinki in that the rights of all human participants are protected through the process of de-identification of data and ethical consideration in the analysis and reporting of data.26Diagnosis was coded as per the WHO guidelines for primary diagnosis,27 as the current outreach data do not allow for clear identification of the presenting issue or diagnosis, as opposed to concurrent ophthalmic comorbidities. Additionally, all cases of trachoma and trichiasis were recoded, even in cases where this was not the primary diagnosis. Where Excel spreadsheets were not available, statistics were sought from written reports. Data were aggregated by country and year. Descriptive statistical analysis was then applied.ResultsIn total, 67 outreach clinics were conducted in seven countries between 2009 and 2013.Outreach data indicate that the mean number of patients seen at outreach clinics appear to be decreasing in Fiji, Samoa and the Solomon Islands (Figure 2). Data were not available for outreach clinics held in Kiribati in 2012 or the Solomon Islands in 2011. Data from PNG stand out for the high mean number of patients per clinic even after accounting for the greater length of PNG outreach clinics.Figure 2: Mean number of patients assessed at outreach clinics by location and year *No data available for clinics held in Kiribati 2012 and the Solomon Islands 2011.Additionally, the number of patients reviewed at PNG outreach clinics appears to be increasing, in contrast to other outreach locations. Notably, outreach clinics in PNG are run differently to those in other locations. PNG does not have locally-based Pacific Eye Institute trained ophthalmic nurses to undertake pre-outreach screening of patients. Therefore, ophthalmic nurses conduct screening simultaneously with medically run outreach clinics and additionally, clinics last for two weeks in PNG, as opposed to one week in all other locations.The percentage of patients whose primary diagnosis is cataract is shown in Table 1. According to these figures, cataract typically contributes to between 40% and 70% of all primary diagnoses at outreach clinics. However, data from PNG in 2013 indicate that the proportion of patients with cataract as a primary diagnosis was considerably lower only 14.1% of all patients reviewed. In contrast, clinics held in Fiji in 2013 had an unusually high proportion of cataract diagnoses. In this year, 94.2% of all patients seen at outreach clinics in Fiji had cataract disease as their primary diagnosis.Table 1: Percentage of outreach patients with cataract as primary diagnosis Clinic locations: 2009 2010 2011 2012 2013 Fiji 65.1 45.4 94.2 Kiribati 59.4 Papua New Guinea 14.1 Samoa 69.2 62.3 60.0 Solomon Islands 64.5 65.1 Tonga 54.9 55.1 Vanuatu 41.1 53.4 48.8 Mean 58.5 56.1 56.7 Note: No data available for any outreach sites from 2009 and 2010.Table 2 illustrates the number of surgeries and the proportion of cataract surgeries performed in each outreach country between 2009 and 2013, according to all available outreach data. This information was not available for all of the outreach clinics during this period. Consequently, this table likely under-represents the true number of surgeries performed at FHFNZ outreach clinics. The proportion of cataract surgeries ranged from as low as 59.0% (Vanuatu, 2009) to 97.1% of all surgeries (Fiji, 2013). It should be noted that the changes observed for Fiji may be related to changes in service provider. In addition, there were no data for PNG and Kiribati.Table 2: Cataract surgery: Annual totals and percentage of all surgeries (in brackets) by location and year Clinic locations: 2009 2010 2011 2012 2013 Total Fiji 89 % unknown 337 (89.6) 387 (92.3) 332 (95.4) 339 (97.1) 1,484 % unknown Kiribati 89 (92.7) 106 (85.5) 195 (88.6) Papua New Guinea 121 (83.4) 183 (85.9) 255 (90.0) 217 (71.0) 776 (82.1) Samoa 105 (82.0) 127 (81.4) 322 (78.5) 167 (85.1) 721 (81.1) Solomon Islands 127 (86.0) 156 (72.0) 126 (60.3) 283 (80.0) 692 (74.6) Tonga 74 (74.7) 207 (77.8) 281 (77.0) Vanuatu 120 (59.0) 188 (62.0) 93 (73.8) 50 (65.8) 116 (66.3) 567 (64.2) Total 209 % unknown 978 (88.8) 946 (83.6) 1,248 (82.1) 1,435 (81.2) 4,816 % unknown Table 3 presents the proportion of all cataract surgeries that were performed on female patients. Data indicate that at most outreach locations, less than half of all cataract surgeries were performed on females. Notably, in 2013 in PNG only 30.4% of all cataract surgeries were undertaken on females.Table 3: Percentage of patients who recieved cataract surgeries who are female Clinic locations: 2009 2010 2011 2012 2013 Fiji 53.9 58.4 63.3 Kiribati 59.8 35.7 Papua New Guinea 30.4 Samoa 45.3 49.0 49.7 Solomon Islands 46.8 45.1 Tonga 45.9 59.6 Vanuatu 48.8 35.7 42.5 Note: No data available for any outreach sites from 2009 and 2010. Table 4 presents the percentage of patients whose primary cause of VI or B is an uncorrected refractive error (URE). Available data indicate that refractive errors contribute a much greater proportion of cases of VI/B at outreach clinics in PNG compared to other locations. In 2013, 60.6% of all patients presenting to outreach clinics in PNG had a primary diagnosis of URE.Table 4: Percentage of outreach clinic patients with uncorrected refractive error as the primary diagnosis Clinic locations: 2009 2010 2011 2012 2013 Fiji 6.8 33.7 0 Kiribati 1.1 Papua New Guinea 60.6 Samoa 2.8 7.3 9.3 Solomon Islands 2.6 2.4 Tonga 0.4 1.7 Vanuatu 21.5 10.3 3.9 Note: No data available for any outreach sites from 2009 and 2010. The proportion of patients at outreach clinics whose primary diagnosis is diabetic retinopathy (DR) is displayed in Table 5. Data are only available from 2011 onwards. DR accounted for less than 4.0% of primary diagnoses at most locations between 2009 and 2011. However, data from Tongan outreach clinics showed a much greater proportion of DR cases: 26.2% in 2012 and 14.2% in 2013.Table 5: Percentage of outreach clinic patients with diabetic retinopathy as the primary diagnosis Clinic location: 2009 2010 2011 2012 2013 Fiji 1.4 0.2 0.3 Kiribati 0.6 Papua New Guinea 0.2 Samoa 3.2 3.7 0.7 Solomon Islands 1.1 0.6 Tonga 26.2 14.3 Vanuatu 0.8 1.7 1.2 Note: No data available for any outreach sites from 2009 and 2010. All patients who were identified as having trachoma or trichiasis in outreach data were recorded (Table 6). This included cases where trachoma or trichiasis did not qualify as the primary diagnosis. Information on the stage of trachoma infection was not available in outreach data. No cases of trachoma or trichiasis were identified in data collected at outreach clinics in Kiribati, Samoa or Tonga. A total of 51 patients with trachoma or trichiasis were identified between 2009 and 2013. The majority of these cases (35 people) were identified at a single outreach clinic in PNG. Notably, 90% of all cases in the Solomon Islands underwent surgical correction for trichiasis.Table 6: Trachoma cases by location: 2011 to 2013 Clinic locations: Number of patients Proportion of cases seen % Female Mean age % with BTR # Fiji 2 0.1 100.0 38.7 0 Kiribati 0 Papua New Guinea 35 51.4 0 Samoa 0 Solomon Islands 10 1.3 70.0 65.3 90.0 Tonga 0 Vanuatu 4 0.6 50.0 69.3 75.0 * Fraction of clinics where trachoma was specifically recorded in outreach reports or where raw data was available for analysis.# Bilamellar tarsal rotation surgical treatment of trichiasis. DiscussionPreventable vision loss remains a persistent public health concern in the Pacific Islands region. Contextual factors, including the vast geographical isolation, chronic shortage of eye nurses and ophthalmologists and, to some extent, a weak primary health sector, contribute to the burden faced by the region.Detailed studies of the epidemiology of eye conditions have been carried out in a small number of PICs, but these are costly and alternative approaches need to be considered. Analysis of routinely collected clinical data cannot replicate the full understanding gained from dedicated population-based eye health surveys. However, as this paper illustrates, using the activities of a single eye health provider can be an efficient means to capture the profile of presenting cases, to understand in broad terms patient demography and to identify hot spots of infectious diseases, for example, trachoma. This method is particularly promising when there is an agreement between local health systems and NGO eye health providers to share data.At present, it is difficult to confidently establish how many organisations are providing eye health services in PICs, the quality of these services, and what impact services are having on the burden of eye disease in the Pacific.This situation is exacerbated by the paucity of detailed ophthalmic epidemiology data from Pacific Island countries. Moreover, maintaining responsive databases is prohibitively expensive and must compete with other health care demands.Data generated at outreach clinics may provide an on-going flow of information to support surveillance of eye health in the region. These records can be used for clinical surveillance to monitor the changes seen in disease presentations over time and across locations, and in this way assist eye health teams to evaluate their work and assist with planning and the prioritisation of resources for outreach services. Outreach eye health services tend to be provided by regional NGOs or by sporadic visiting international medical teams, possibly leading to minimal follow-up and no clear accountability to national or regional agencies. Therefore, linking the various individual and NGO eye health service data to national (and regional) health information systems may greatly enhance data quality.A commitment to the collection and open reporting of outreach data by eye health service providers in the Pacific will allow more accurate prediction of emerging trends in the ophthalmic health of PICs. This is of great importance for monitoring the changes in disease presentations, particularly as PICs transition toward a greater burden of non-communicable diseases.27-30 It is likely in the future there will be greater numbers of patients presenting with cataract, DR31,32 and hypertensive retinopathy.33,39 Eye health teams can also contribute key data to determine access to eye health services for vulnerable groups, such as women, children, elderly and ethnic minority groups. This information provides in-country government agencies and visiting ophthalmology teams a clearer profile of groups with poorer access to eye health services and supports the development of strategies to address disparities.Data collected from Pacific outreach eye health clinics between 2009 and 2013 indicate presentations at outreach clinics decreased in three out of the seven counties surveyed (Fiji, Samoa and the Solomon Islands). Several explanations are viable here: a decline in the number of patients due to a diminishing backlog of cataract-induced VI/B; changes to the number and/or type of clinicians at outreach clinics; or cataract prevalence may remain high, but patients are not presenting to outreach clinics for treatment. Alternatively, the changing disease profile in the region may mean patients are not presenting as their conditions are largely asymptomatic, undermining the reason and certainly the urgency of presenting to an outreach clinic.Of note, our data also showed high numbers of URE cases were reported in PNG a condition associated with poor access to primary eye health services. Tongan clinics are notable for the high number of DR cases reported, consistent with the high prevalence of diabetes in that country.28 Also, our findings indicate that between 2009 and 2013, in most locations, more men than women underwent cataract surgery, although most studies find a higher prevalence of cataract amongst women in developing countries.28-32Likewise, clinical surveillance is critical for providing an estimate of the burden of trachoma in PICs in the absence of up-to-date survey data on the prevalence and effects of trachoma in PICs. Collaborative use of clinical data from outreach clinics could provide an insight into where trachoma infections are prevalent or where a greater proportion of infections are progressing to VI/B. Eye health teams should also consider using negative reporting to ensure that all cases of trachoma are identified, given the drive to see blinding trachoma eliminated globally34-36 and the current lack of information on its prevalence and effects in the Pacific.22,35Our analysis of outreach data provides evidence of the importance of investing in robust and reliable data collection systems that prioritise accountability and transparency. We argue that bold innovation and collaboration in data collection, reporting and sharing mechanisms are essential in resource scarce settings. In the absence of this, we will continue to be prone to speculation about the shifting burden of eye health and disease in the region.ConclusionAnalysis of one eye health service providers data from outreach clinics across seven PICs from 2009 to 2013 indicates a reduction in the number of patients presenting for treatment. This trend, and others reported here, should be interpreted cautiously due to concern about data completeness and quality. Also, these data relate to those who present after (in most countries) screening by the eye care nurses. We have no information on those who may suffer from eye health problems, but do not present to clinics.A commitment to quality data management and surveillance strategies for all eye health outreach programmes is vital to ensure resources are reaching those who need it, that health equity lies at the core of the efforts to deliver eye health services and finally, that innovation and collaboration are essential components of Pacific regional eye health initiatives.